股票日内价格过度反应的统计分析

发布时间：2018-04-10 23:29

本文选题：过度反应 + 日内价格　；参考：《华东师范大学》2012年硕士论文

【摘要】：在现代金融理论,与有效市场理论(EMH)相背离的最具代表性的“异象”便是过度反应。首先对过度反应现象进行研究的是DeBondt和Thaler(1985),他们研究发现,美国股票市场存在价格的长期反转现象。过度反应的发现挑战着传统金融理论的权威。但是,传统金融学仅仅把过度反应看作“异象”,比如：Fama和French(1998)认为超常收益来源于风险的补偿,并尝试引入了新的因子来刻画风险。传统金融理论的解释并不能令人信服。与此同时,行为金融学给出的解释和相关的模型似乎更具有说服力。比如：Barberis, Shleifer和Vishny的BSV模型,Daniel, Hirsheifer和Subramanyam的DHS模型,Hong和Stein(1999)的HS模型。各国学者针对证券市场是否存在过度反应进行了大量实证检验,所得出的结论却存在着较大的分歧。其中最重要的一个原因就是定义过度反应的标准和检验的方法不同。文本主要讨论了定义过度反应的比较标准和检验方法的改进。本文第一章主要介绍了过度反应的研究背景,研究现状和存在的问题。第二章主要采用定量的方法,基于股票日内价格,在股票日内价格服从几何布朗运动的假定下,构造了一个新的衡量波动率的统计量；并以此为基础,从Hausman检验的角度构建检验日内价格过度反应的检验统计量并得到其理论分布。第三章分别讨论了波动率变化(GARCH),漂移率变化(AR),波动率和漂移率均变化(AR-GARCH)的情况下检验统计量的稳健性并给出相应的修正检验统计量。第四章运用此统计量检验了全球不同市场的过度反应现象,并通过实际数据也验证了改进检验统计量的稳健性和适用性。最后,尝试探讨了基于过度反应检验统计量的交易策略。
[Abstract]:In modern financial theory, the most representative "vision" that deviates from efficient market theory is overreaction.First, the overreaction phenomenon is studied by DeBondt and Thalerian 1985. They found that there is a long-term price reversal in the American stock market.The discovery of overreaction challenges the authority of traditional financial theory.However, the traditional finance only regards overreaction as a "vision", such as: Fama and French 1998, that the extraordinary income comes from the compensation of risk, and tries to introduce new factors to depict the risk.The explanation of traditional financial theory is not convincing.At the same time, behavioral finance's explanations and related models seem more persuasive.For example, Shleifer and Vishny's BSV model, Hirsheifer's and Subramanyam's DHS model and Hong's and Stein's 1999 are HS models.Scholars from all over the world have made a lot of empirical tests on whether there is overreaction in the securities market, but the conclusions are quite different.One of the most important reasons is that the criteria for defining overreaction differ from the test.This paper mainly discusses the comparison standard of defining overreaction and the improvement of test method.The first chapter mainly introduces the research background, research status and existing problems of overreaction.In the second chapter, based on the stock intraday price, a new statistical measure of volatility is constructed under the assumption of geometric Brownian motion.From the point of view of Hausman test, the test statistic of intraday price overreaction is constructed and its theoretical distribution is obtained.In the third chapter, we discuss the robustness of the test statistics under the condition of the variation of volatility, the variation of drift rate, the variation of volatility and the variation of drift rate, respectively, and the corresponding revised test statistics are given.In chapter 4, we use this statistic to test the overreaction phenomenon in different markets around the world, and verify the robustness and applicability of the improved test statistics through the actual data.Finally, the trading strategy based on overreaction test statistics is discussed.
【学位授予单位】：华东师范大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F224;F832.51

【参考文献】